To the Editor: The meta-analysis by Dr Barone and colleagues1 showed that patients diagnosed with cancer who have preexisting diabetes are at increased risk for all-cause mortality compared with those without diabetes. We agree with this conclusion but have 2 comments. First, the study by Kouvaris et al2 included only 30 cancer patients with diabetes and 50 patients in total. Therefore, we wonder whether in Figure 2 the random-effects weight and the large data marker are correct.
van de Poll-Franse L, Coebergh JW. Long-term Mortality in Cancer Patients With Preexisting Diabetes. JAMA. 2009;301(15):1541. doi:10.1001/jama.2009.488